Immune responses to foreign antigens such as those found in bacteria and virus protect from and eliminate infections. However, aberrant immune responses can cause allergic diseases and autoimmune diseases. Immune responses to foreign, sometimes innocuous, substances such as pollen, dust mites, food antigens and bee sting can result in allergic diseases such as hay fever, asthma and systemic anaphylaxis. Immune responses to self-antigens such as pancreatic islet antigens and cartilage antigens can lead to diabetes and arthritis, respectively. The hallmark of the allergic diseases is activation of CD4 T cells and high production of IgE by B cells, whereas the salient feature of autoimmune diseases are activation of CD4 T cells and over production of inflammation cytokines. The current therapies have been focused on the treatment of symptoms of allergy and autoimmune diseases and do not prevent the development and progression of the diseases.
CTLs are derived from resting naïve CD8 T cells and recognize antigenic peptides presented by Major Histocompatibility Complex (MHC) class I molecules. When resting CD8 T cells encounter antigenic peptides/MHC complex presented by professional antigen presenting cells, CD8 T cells will be activated and differentiated into armed CTL. Upon recognition of peptide/MHC complexes on the target cells, the antigen specific CTL will deliver a lethal hit and lysis the antigen-expressing target cells, such as virus infected target cells or tumor cells. Activation of naive T cells in vivo is controlled by multiple receptor-ligand interactions between T cells and professional APC such as dendritic cells (R. M. Steinman, Annu. Rev. Immunol. (1991) 9:271-296). It is generally accepted that two signals are required for activation of naive T cells (C. A. Janeway and K. Bottomly, Cell (1994) 76:275-285). Signal 1 is induced by the interaction between TCR and MHC/peptide complexes (R. N. Germain, Cell (1994) 76:287-299) and is aided by binding of CD4/CD8 co-receptors to non-polymorphic regions of MHC class II/I molecules, respectively (M. C. Miceli and J. R. Parnes, Adv. Immunol. (1993) 53:59-122). Signal 2 is qualitatively different from Signal 1 and is delivered via T cell costimulatory molecules interacting with complementary ligands on APC, e.g. through CD28 interaction with B7 (P. S. Linsley and J. A. Ledbetter, Annu. Rev. Immunol. (1993) 11:191-212; Lenschow et al., Annu. Rev. Immunol. (1996) 14:233-258). Signals 1 and 2 function synergistically and trigger a series of signaling events which ultimately induce T cells to proliferate, produce cytokines and differentiate into effector cells (Mueller et al., Annu. Rev. Immunol. (1989) 7:445-480; A. Weiss and D. R. Littman, Cell (1994) 76:263-274). The relationship between Signals 1 and 2, however, is unclear.
Although a variety of molecules have been reported to have costimulatory function, particular attention has been focused on costimulation delivered via CD28-B7 interaction (R. H. Schwartz, Cell (1992) 71:1065-1068). CD28 is a molecule with a single Ig like domain and is constitutively expressed as a homodimer on T cells (P. S. Linsley and J. A. Ledbetter, (1993) supra). Through its interaction with either B7-1 or B7-2 molecules on APCs, CD28 molecules are thought to transduce unique signals that stimulate T cell to produce growth-promoting cytokines such as IL-2 (June et al., Immunol. Today (1994) 15:321-331), to upregulate expression of survival factors such as Bcl-XL (Boise et al., Immunity (1995) 3:87-98) and to prevent anergy induced by Signal 1 alone (R. H. Schwartz, Curr. Opin. Immunol. (1997) 9:351-357).
Another pair of molecules that has an important role in T cell activation is LFA-1/ICAM-1 (Van Seventer et al., J. Immunol, (1990) 144:4579-4586). ICAM-1 belongs to the Ig gene superfamily and has five Ig C like domains in the extracellular regions; it is expressed on both hemapoietic and nonhemapoietic cells. The receptor for ICAM-1 on T cells is LFA-1 (CD11/CD18), which belongs to the b2 integrin family (T. A. Springer, Cell (1994) 76:301-314). The interaction of LFA-1 with ICAM-1 has potent costimulatory function on T cells (Shimizu et al., Immunol. Rev. (1990) 114:109-143), although opinions vary on whether this function reflects a separate signaling pathways or increased adhesion between T cells and APC (Damle et al., J. Immunol. (1993) 151:2368-2379; Bachmann et al., Immunity (1997) 7:549-557).
In addition to B7 and ICAM-1 molecules, several other molecules on APCs, including CD70 (Hintzen et al., J. Immunol. (1995) 154:2612-2623) and heat-stable antigen (HSA) (Liu et al., J. Exp. Med. (1992) 175:437-445), can exert quite potent costimulatory function through their interaction with their respective ligands on T cells. The implication is that T-APC interaction is highly complex and involves multiple interactions between complementary sets of molecules on T cells and APCs. The interaction of each set of molecules could trigger specific signals which induce different cellular events. The combination of the different signals may act synergistically for optimal T cell activation and determine the final fate of T cells. Alternatively, the function of costimulation molecules may be redundant and the signals induced by each set of costimulation molecules are additive. The requirement for each set of costimulation molecules will be influenced by the strength and characteristics of Signal 1.
In considering these two possibilities, it is important to understand the minimal requirements for stimulating naive T cells. Studies with CD28−/− mice indicated that CD28-B7 interaction is highly important in some situations, but not in others (Shahinian et al., Science (1993) 261:609-612). Likewise, the requirement for LFA-1/ICAM interaction in primary responses is not an invariable finding (Shier et al., J. Immunol. (1996) 157:5375-5386).
CD8 T cells recognize antigenic peptides derived mainly from virus proteins and proteins expressed on tumor cells. However, it has recently been reported that newly synthesized proteins are preferentially processed by antigen-processing machinery (Schubert et al., Nature, (2000) 404:770-774). Upon activation, immune cells have acquired the ability to synthesize a number of new proteins, it is possible that IgE producing B cells and activated CD4 T cells would present a different sets of peptide/MHC complexes than the non-IgE producing cells and resting CD4 T cells. These peptides/MHC complexes presented on IgE producing B cells and activated CD4 T cells would be able to be recognized by CD8 T cells. Thus, CTL specific for these peptides/MHC complexes would be able to treat allergy and autoimmune diseases. However, a number of tolerance mechanisms have been able to prevent the activation the CD8 T cells towards self-antigens in vivo.
CD8 lymphocytes (CTLs) are the arm of adaptive immunity responsible for the recognition and elimination of infected cells, tumor cells, and allogeneic cells. Once primed, CTL can recognize their target antigen on a wide variety of cells and accomplish their function by lysing the target cell and/or secreting cytokines like TNF-alpha, or IFN-gamma.
Presentation of antigen to CD8+ CTL (cytotoxic T lymphocytes) occurs in the context of MHC class I molecules (MHC-I), while presentation of antigen to CD4+ HTL (helper T lymphocytes) occurs in the context of MHC class II molecules.
Efficient induction of CD4+ T cell requires that the T cells interact with antigen presenting cells (APC) i.e. cells that express MHC class II and co-stimulatory molecules. APC are dendritic cells, macrophages and activated B cells. Although nearly all nucleated cells express MHC-I, naive CTL also require presentation of antigen (Ag) by bone marrow-derived APC for efficient priming (Dalyot-Herman et al., J. Immunol., 165 (12):6731-6737). Dendritic cells are highly potent inducers of CTL responses (J. Bancherean and R. M. Steinman, Nature, (1998) 392:245-252) and are thought to be the principal APC involved in priming CTL. Once primed, CTL can recognize their cognate Ags on a wide variety of cells and respond by lysing the target cell and/or secreting cytokines.
Although bone marrow-derived APC are required to efficiently prime CTL responses (P. J. Fink and M. J. Bevan, Exp. Med. (1978) 148:755-766) activated CTL are readily able to recognize and respond to Ag presented by a wide variety of cells. Induction of tumor- or viral-specific CTL immune responses in vivo have been shown to be dependent on bone marrow derived antigen-presenting cells (Paglia et al., J. Exp. Med. (1996) 183 (1):317-322; Labeur et al., J. Immunol. (1999) 162 (1):168-175). It is generally accepted that bone marrow derived APC, through mechanisms unique to these cells, take up cellular antigens either in the form of soluble antigen, associated with chaperone molecules or by phagocytosis.
It has long been demonstrated that responses to cellular antigens are dependent on help delivered by CD4+ T cells. It has also been shown that the cellular antigen had to be presented on the same APC for recognition by the CTL and the HTL. The nature of this help has been interpreted as a need of IL-2 necessary for CTL expansion. Recent studies have shown that this help results from the activation of dendritic cells by HTL and is mediated via CD40-CD40L interaction (S. R. Clarke, J. Leukocyte Bio. (2000) 67 (5):607-614).
A likely scenario for the induction of a CD8 mediated immune response to a cellular antigen (derived from a tumor cell or an infected cell) is therefore the following: dendritic cells acquire antigens derived from tumor or infected cells. Interaction of DC-antigen with CD4 cells enable the DC to activate the CD8 cells.